The Machine That Defied Physics Long Enough to Cash a Check: America's Most Embarrassing Patent
The United States Patent and Trademark Office has a rule about perpetual motion machines. It's been on the books since 1911, and it's about as close to an absolute policy as the federal government ever gets: before a patent examiner is required to seriously consider any device claiming to produce more energy than it consumes, the inventor must submit a working model.
This rule exists because the USPTO has been burned before. Many times. Across many decades. By many enthusiastic men with many impressive-looking diagrams and a spectacular variety of excuses for why the working model wasn't quite ready yet.
And yet, somehow, in the 1970s, an inventor managed to navigate around this safeguard and walk out of the patent system with an official U.S. patent for a device that was, by any honest reading of the laws of thermodynamics, physically impossible.
The story of how that happened is a small masterpiece of bureaucratic failure, wishful thinking, and the eternal human desire to believe that this time, finally, someone has actually beaten the universe.
The Machine and the Man
The inventor — an eccentric engineer with a genuine background in mechanical systems and a long history of unconventional ideas — had been working on his concept for years before filing. His device was built around a clever arrangement of magnets, flywheels, and precisely calibrated mechanical components that he claimed could sustain and amplify their own motion indefinitely, generating usable electrical output in the process.
On paper, the design was sophisticated enough to be confusing. That was, in retrospect, a significant part of its success. The patent application ran to dozens of pages of technical description, filled with the kind of dense mechanical terminology that requires genuine expertise to evaluate. The drawings were meticulous. The language was careful. The inventor had learned, whether deliberately or through instinct, that the best way to get a complicated false claim past a reviewer is to make it complicated enough that the reviewer isn't entirely sure what they're reviewing.
The patent examiner assigned to the application was not a physicist. This mattered.
How the System Failed
The USPTO's working model requirement for perpetual motion devices is a strong protection, but it has a weakness: it applies specifically to devices claiming perpetual motion. If an inventor frames their application carefully enough — emphasizing efficiency gains, novel mechanical arrangements, and energy recovery rather than the net energy output — the trigger for the working model requirement may never formally activate.
That appears to be roughly what happened here. The application was framed in terms of mechanical innovation and energy efficiency rather than as an explicit perpetual motion claim. The examiner, working through a substantial backlog with limited time for deep technical review, processed it through the standard examination pipeline.
The patent was granted.
For the inventor, this was not just a piece of paper. A U.S. patent is a powerful marketing document. It signals, to anyone who doesn't read the fine print, that the federal government has reviewed and approved your invention. Investors who might have dismissed a perpetual motion claim from an unknown engineer looked considerably more interested when that engineer showed up with an official patent certificate.
Money started moving.
The Investors and the Demonstrations
The inventor was, by multiple accounts, a genuinely compelling salesman. He gave demonstrations of his device to potential investors, and the demonstrations were impressive — the machine ran, produced measurable output, and behaved in ways that were difficult to immediately explain away.
What the demonstrations didn't reveal, and what took investigators considerable time to establish, was that the device was not operating the way its inventor claimed. The energy accounting was wrong. Measurements were being taken selectively. Components that appeared to be passive were, under careful examination, doing work in ways that weren't being disclosed.
None of this was immediately obvious. The device was complex. The inventor was knowledgeable. And the investors — like most people confronted with something they don't fully understand but very much want to believe — extended a great deal of benefit of the doubt.
By the time the scientific and regulatory communities caught up with the claims, a meaningful amount of money had changed hands.
Thermodynamics Gets the Last Word
Physicists who eventually examined the device were not gentle in their assessment. The First and Second Laws of Thermodynamics are not suggestions. Energy cannot be created from nothing. Entropy does not run backward. No arrangement of magnets and flywheels, however clever, can produce a net energy gain from a closed system. These are not theoretical positions subject to revision by a creative engineer with a good patent attorney — they are among the most thoroughly tested and consistently confirmed principles in the history of science.
Once serious technical scrutiny was applied, the device's actual performance collapsed immediately. The apparent energy gains evaporated under controlled measurement conditions. The machine was, in the end, exactly what the laws of physics had always predicted it would be: a very expensive way to convert input energy into slightly less output energy, with the remainder lost to heat and friction.
The patent was eventually challenged and its claims formally discredited. The USPTO, for its part, conducted an internal review of how the application had cleared examination and tightened its procedures for evaluating energy-related patents.
As for the inventor, he pivoted. In the years following the collapse of his perpetual motion enterprise, he filed additional patents for devices in adjacent territory — each one carefully worded, each one claiming something just plausible enough to avoid immediate dismissal. Whether any of them worked is a question that probably answers itself.
The Eternal Appeal of Something for Nothing
Perpetual motion machines have been fooling people since at least the 12th century, when a monk named Bhaskara II described a self-turning wheel that would, he claimed, rotate forever. They have never worked. Not once. Not even a little bit.
And yet the applications keep coming. The USPTO reportedly receives a small but steady stream of perpetual motion and over-unity device applications every single year, from inventors who are convinced that they have finally solved the problem that has defeated every scientist and engineer in human history.
Most of them are rejected. Some of them, if the framing is clever enough and the examiner is tired enough, make it further than they should.
The universe, to its credit, always settles the question eventually. It just occasionally lets someone cash a check first.